Kenya Prep Rollout Analysis Update July 2019

Kenya PrEP Rollout Analysis Update

July 2019 Overview

• In 2015, the OPTIONS Consortium supported NASCOP to develop an analysis of possible oral PrEP rollout scenarios at the county level in Kenya. That analysis was based on two key sources of data – HIV incidence and prevalence estimates and estimates of key populations at the county level.

• In July 2019, the OPTIONS Consortium updated this analysis with the new 2018 Kenya HIV Estimates Report data. This updated analysis is included in the following slides.

• The analyses on the following slides include a snapshot of HIV incidence across counties, a framework to inform PrEP rollout, specific data on high priority oral PrEP client populations (serodiscordant couples and adolescents), and initial data from four counties on how current oral PrEP rollout compares to the rollout analysis.

• This analysis will be updated again later in 2019 as additional data becomes available on (1) updated key population estimates by county and (2) current PrEP rollout in additional counties.

• Please contact Jordan Kyongo at LVCT Health ([email protected]) with any questions.

2 Kenya County HIV Incidence

HIV Incidence by County, 2018

High incidence cluster 1 Incidence rates above 0.18 (national average) (Homa Bay, Siaya, Kisumu, Migori, Busia, Kiambu, Nairobi, Murang’a, Vihiga, Mombasa)

2 Medium incidence cluster Incidence rates of 0.1-0.18 (Kakamega, Nyandarua, Nyeri, Kitui, Taita Taveta, Kilifi, Kirinyaga, Kwale, Makueni, Kisii, Machakos, Nyamira, Bungoma, Isiolo, Lamu, Tharaka – Nithi, Embu)

3 Low incidence cluster Incidence rates below 0.1 (Meru, Trans Nzoia, Kajiado, Uasin Gishu, Nakuru, Turkana, Kericho, Narok, Laikipia, Marsabit, Tana River, Bomet, Nandi, Samburu, Baringo, Elgeyo-Marakwet, West Pokot, Garissa, Mandera, Wajir)

Source: Kenya HIV Estimates Report 2018 (Kenya Ministry of Health, October 2018) 3 Kenya New Adult HIV Infections

Proportion of National Adult New HIV Infections by Cluster, 2018

Total # of New Incidence # of Population Infections Cluster Counties (15+) (15+) 11% High incidence High counties 1 incidence 10 8.7M 26K

Medium incidence 30% counties Medium 59% 2 incidence 17 8.4M 13K

Low incidence counties Low 3 incidence 20 9.1M 5K

PrEP delivery should be prioritized in the high incidence county cluster

Source: Kenya HIV Estimates Report 2018 (Kenya Ministry of Health, October 2018) 4 County PrEP Rollout Analysis | Overview

Two-Step Delivery Approach Framework

1 HIV incidence (rate and absolute 2 Size of key populations (FSW, MSM) determines number of new HIV infections) how a county should rollout PrEP. determines a county’s need for investment in new HIV prevention Counties with HIV incidence driven by key solutions including oral PrEP and populations should consider a targeted rollout prioritizes counties for PrEP rollout. to those groups while counties with low key populations but high HIV incidence should Counties with higher HIV incidence consider rollout to the general population, are higher priority for PrEP rollout. including serodiscordant couples, adolescent girls & young women, and bridging populations

HIV Incidence Source: Kenya HIV Estimates Report 2018 (e.g., fisherfolk). (Kenya Ministry of Health, October 2018)

Source: FSW, MSM, PWID estimates, MARPS, 2012

Population-Driven HIV Incidence Generalized HIV Incidence

Counties are mapped to this framework in the following slides

Sources: Informed by Avenir, PrEP for Adolescent Girls and Young Women in Kenya, Preliminary Results Presentation, October 2016 5 County PrEP Rollout Analysis | All populations

9 Size of bubble • PrEP is a priority for counties with high indicates HIV incidence at the top of the chart number of new • PrEP rollout can be more focused on HOMA BAY infections key populations for counties on the (2017) SIAYA left of the chart • PrEP rollout to the general population KISUMU (including AGYW) will be important for MIGORI counties on the right of the chart BUSIA

NAIROBI KIAMBU NYERI VIHIGA MURANG’A MOMBASA NYANDARUA KAKAMEGA National average TAITA TAVETA MACHAKOS KITUI HIV incidence (1.8) KWALE KILIFI KISII MAKUENI ISIOLO BUNGOMA KIRINYAGA THARAKA-NITHI NYAMIRA TRANS NZOIA LAMU MERU EMBU

(New HIV infections per 1,000 population) 1,000 per infections HIV (New KAJIADO KERICHO TURKANA NAKURU MARSABIT UASIN GISHU TANA RIVER NAROK BOMET NANDI WEST POKOT LAIKIPIA ELEGEYO-MARAKWET 0 SAMBURU BARINGO HIV incidence Population-Driven HIV Incidence (Significant key population presence) Generalized HIV Incidence (Low key population presence) 6 County PrEP Rollout Analysis | Serodiscordant couples

9 Size of bubble • With high rates of HIV prevalence, indicates Homa Bay, Siaya, Migori and Kisumu HOMA BAY number of new would benefit from provision of PrEP infections to sero-discordant couples SIAYA • Busia and Nairobi also have relatively Counties with high rates of HIV prevalence and over 10% HIV KISUMU should ensure PrEP access for sero- MIGORI prevalence discordant couples BUSIA Counties with 5 – 10 % HIV prevalence

9 Size of bubble indicates • Adolescents account for a significant number of new portion of Kenya’s HIV incidence – HOMA BAY infections particularly in the shaded counties Counties where • In these counties, PrEP access for SIAYA adolescents (age 15 – 24) adolescents, via adolescent-friendly account for over 40% of delivery channels, will be important to KISUMU new HIV infections drive progress on HIV prevention MIGORI BUSIA Counties where adolescents (age 15 – 24) account for 30 – 40% of new HIV infections

9 Size of bubble indicates • Current data from three counties number of new demonstrates that PrEP rollout has infections HOMA BAY largely followed the implications of this analysis SIAYA • Some counties with high adolescent KIAMBU – Rollout may need to be expanded to HIV incidence should further expand KISUMU populations beyond key populations (e.g., PrEP access for adolescents MIGORI serodiscordant couples, AGYW) as Kiambu has BUSIA SIAYA – Rollout in-line with expectations, with generalized HIV incidence* focus on serodiscordant couples and AGYW New HIV infections: 2,783 New HIV infections: 3,419 Current clients on PrEP: 1,665 Current clients on PrEP: 4,041 - 30% Serodiscordant Couples - 9% AGYW NAIROBI - 70% Serodiscordant Couples KIAMBU - 18% AGYW - 40% Key PopulationsNYERI VIHIGA MURANG’A MOMBASA - 3% Key Populations KAKAMEGA NYANDARUA - 7%TAITA General TAVETA Population MACHAKOS KAKAMEGA – Rollout mayKITUI need to be further KWALE expanded to adolescents, as they account for a high KILIFI KISII MAKUENI LAMU proportion of new HIV infections in the county BUNGOMA NYAMIRA KIRINYAGA THARAKA-NITHI NewEMBU HIV infections:TRANS 2,197 NZOIA ISIOLO MERU Current clients on PrEP: 588 TURKANA (New HIV infections per 1,000 population) 1,000 per infections HIV (New -KAJIADO80% SerodiscordantKERICHO Couples NAKURU - <1% AGYW NAROK TANA RIVER MARSABIT UASIN GISHU - 4% Key PopulationsLAIKIPIA BOMET NANDI WEST- 15%POKOT General Population ELEGEYO-MARAKWET 0 SAMBURU BARINGO HIV incidence Population-Driven HIV Incidence (Significant key population presence) Generalized HIV Incidence (Low key population presence) * Key population estimates were last conducted in 2012, therefore this data may be outdated of the situation has changed in Kiambu 9 APPENDIX

10 Analysis data

Adult* Adult* HIV Adult* HIV Youth** Adult* Adult* HIV Adult* HIV Youth** Adult (15+) Key Adult (15+) Key HIV Incidence New New HIV HIV Incidence New New HIV County Population Populations County Population Populations Prevalence (# per 1,000 Infections Infections Prevalence (# per 1,000 Infections Infections (100,000s) (MSM + FSW) (100,000s) (MSM + FSW) (%) people) (#) (#) (%) people) (#) (#) BARINGO 3.7 - 1.3% 0.3 102 44 MARSABIT 1.7 - 1.4% 0.5 87 27 BOMET 4.9 550 1.9% 0.4 184 80 MERU 8.7 3,163 2.4% 0.9 813 251 BUNGOMA 7.9 4,195 3.2% 1.3 999 338 MIGORI 5.4 2,945 13.3% 5 2,382 1,143 BUSIA 4.3 3,327 7.7% 3.1 1,283 434 MOMBASA 8.2 10,070 4.1% 1.9 1,490 562 ELEGEYO-MARAKWET 2.5 - 1.6% 0.3 83 36 MURANG'A 7.1 626 4.2% 2 1,350 376 EMBU 3.5 1,061 2.8% 1.1 363 112 NAIROBI 28.1 29,190 6.1% 2.2 6,499 2,587 GARISSA 2.6 - 0.8% 0 0 0 NAKURU 11.9 5,568 3.4% 0.7 860 374 HOMA BAY 5.8 1,334 20.7% 8.2 3,858 1,852 NANDI 5.3 - 2.0% 0.4 221 96 ISIOLO 0.9 - 3.2% 1.3 106 33 NAROK 5.4 580 2.7% 0.6 317 138 KAJIADO 5.1 1,590 3.9% 0.8 432 187 NYAMIRA 3.9 974 4.2% 1.4 528 253 KAKAMEGA 9.9 3,719 4.5% 1.8 1,761 596 NYANDARUA 4.2 826 3.5% 1.8 711 198 KERICHO 5.4 1,116 2.9% 0.6 304 132 NYERI 5.3 997 3.7% 1.8 952 265 KIAMBU 13.0 4,913 4.0% 2.2 2,623 730 SAMBURU 1.4 - 1.8% 0.4 53 23 KILIFI 8.2 5,316 3.8% 1.6 1,183 446 SIAYA 5.4 2,767 21.0% 7.7 3,419 1,641 KIRINYAGA 4.2 759 3.1% 1.6 644 179 TAITA TAVETA 2.3 1,530 4.1% 1.7 366 138 KISII 7.4 4,489 4.4% 1.5 1,052 505 TANA RIVER 1.7 - 1.3% 0.5 80 30 KISUMU 6.4 5,671 16.3% 6.3 3,396 1,630 THARAKA-NITHI 2.4 737 3.2% 1.2 286 88 KITUI 5.8 794 4.5% 1.7 970 299 TRANS NZOIA 5.5 828 4.3% 0.9 503 218 KWALE 4.8 1,369 3.8% 1.6 691 261 TURKANA 5.9 724 3.2% 0.7 403 175 LAIKIPIA 2.9 558 2.7% 0.5 161 70 UASIN GISHU 6.7 2,537 3.9% 0.8 560 243 LAMU 0.8 - 3.0% 1.3 95 36 VIHIGA 3.4 2,900 5.4% 2 663 224 MACHAKOS 7.2 4,002 3.8% 1.4 1,019 314 WAJIR 2.6 - 0.1% 0 0 0 MAKUENI 5.3 2,012 4.2% 1.6 832 257 WEST POKOT 3.1 1,012 1.6% 0.3 104 45 MANDERA 3.9 - 0.2% 0 0 0 * Adults aged 15 – 49 ** Youth aged 15 – 24 Sources: Kenya HIV Estimates Report 2018 (Kenya Ministry of Health, October 2018); FSW, MSM, PWID estimates (MARPS, 2012) 11